Preparation of LTS Setups¶
Set-up and calibration of a collection system model¶
This step is identical as for any other collection system model. I.e. the model shall be set-up and calibrated, following a good modelling practice and calibration principles.
The model stability shall be tested by simulating high-intensity rainfall events (historical or synthetic) so that an appropriate simulation time step for the LTS simulation can be determined: a shorter time step contributes to the simulation stability but may compromise the simulation efficiency.
Preparation of the LTS simulation¶
Preparing the LTS simulation requires to choose the simulation period, prepare the LTS catchment boundary time series and define the LTS catchment model boundary condition.
A LTS simulation normally covers long historical periods, ranging from several years to several decades (10-30 years). Duration of the simulation period may be limited by the purpose of the LTS study and, often, by the limited availability of reliable historical rainfall time series. A longer simulation period is to be preferred, as it creates a better foundation for the statistical analysis.
The applied rainfall time series should be as complete as possible. Application of a time series with frequent drop-out events may generate misleading statistical results.
If the rainfall time series includes events known to be recorded incorrectly or if such events have been removed from the time series, they should be replaced by correctly recorded rainfall at the given time at a nearby rain gauge (if available).
Any selected rainfall time series (one or several) shall be included as the catchment model rainfall boundary condition. If the hydrological simulation includes RDI model, then the potential evapotranspiration time series for the model area, covering the LTS simulation period shall also be provided as boundary condition. Similarly, where relevant, air temperature time series must be provided to support the simulation of snow accumulation/snow melt processes.
Definition of job list criteria¶
At least one job list criterium shall be specified. Otherwise, any number of job list criteria may be specified.
The specified START threshold shall be defined so that any event of relevance for the actual analysis is included in the LTS job list. A too high threshold value may cause important loss of information (i.e. exclusion of events that contribute to the statistics. On the other hand, a too low threshold means inclusion of many, potentially insignificant events in the LTS job list, which may compromise the efficiency of the LTS simulation.
A good guideline for the threshold value may be intensity of specific runoff (\(LT^{-1}\)), i.e. runoff per unit area. E.g. if it is known that overflows are likely to occur if the specific runoff exceeds certain intensity, the threshold may be calculated based on the total catchment area connected to the specified location.
"Duration" depends greatly on the specified location (size of the connected catchments). Small catchments may generate very short runoff peaks associated with spikes in recorded rainfall. Specifying a longer "duration" may eliminate such events from the LTS simulation.
The specified END threshold should reflect a situation when the inflow is low enough to be considered as insignificant for the processes of interest. Usually, it can be defined as a fraction of the START threshold (e.g. 10%). To ensure that the delayed hydraulic effects in the network are included in the simulation, appropriate "Duration" is specified. This depends on the size of the system, emptying time of the retention basins, etc.
Note
A better and a more precise way to control the END of the event simulation is by means of runtime criteria.
Definition of runtime stop criteria (optional)¶
It is recommended to apply run time stop criteria, as this is much more precise method to control the end of the event simulation than job list stop criteria. Precisely specified runtime stop criteria will result in the shortest total simulation time, without any loss of important information due to prematurely ending event simulations.
Preparation initial conditions¶
The next step consists in preparing hotstart files and defining initial conditions for LTS simulation events (optional).
Realistic initial conditions for the LTS simulation must be provided in order to achieve correct results. The following cases should be considered:
- Storm drainage system loaded by a surface runoff model. It is assumed that the system is empty in dry weather. The runoff model generates intermittent runoff loads associated with rainfall events. Appropriate initial conditions for such a setup are "empty system", i.e. no hot start file is required.
- Wastewater and/or combined drainage system with some dry weather loads and runoff loads generated by a surface runoff model. In dry weather, the system carries dry weather loads, typically wastewater. At the beginning of rainfall event, hydrological load is zero. Appropriate initial conditions for such a setup is with a single hot start file, created with dry weather loads only. The hot start file shall include at least one day of dry weather flows, excluding the filling phase. This can be achieved either running the DWF hot start simulation over two or several days, until a fully stationary (repetitive) situation is established.
- Wastewater and/or combined drainage system with some dry weather loads and runoff loads generated by RDI runoff model (continuous infiltration). In dry weather, the system carries dry weather loads, typically wastewater, and some infiltration load. At the beginning of rainfall event, the hydrological load (i.e. infiltration) may have a different scale, depending on previous hydrological events, season, etc. Appropriate initial conditions for such a setup are established with several hot start files, created with dry weather loads and a constant infiltration component. The hot start files shall cover the expected range of possible infiltration in uniform intervals. The hot start files shall include at least one day of dry weather flows, excluding the filling phase. This can be achieved either running the DWF hot start simulation over two or several days, until a fully stationary (repetitive) situation is established.
When a hotstart file (or files) is/are prepared, initial conditions for the LTS event simulations shall be specified in the dedicated "LTS initial conditions editor".
Review and modification (optional) of LTS global parameters¶
Default values for LTS global parameters represent a reasonable choice for the most case. It is, however, a good idea to review the global parameters and consider modifications, if it deems appropriate for the actual LTS setup.
Definition of LTS outputs¶
Depending on the specific requirements, LTS output files should be specified and included in the actual simulation setup.